80legs 0.1.0 → 0.2.0

data/VERSION

@@ -1 +1 @@
-0.1.0
+0.2.0

data/lib/eighty_legs/eighty_format.rb

@@ -1,16 +1,27 @@
 module EightyLegs
   class EightyFormat
     def initialize(filename_or_io)
-      @io = if filename_or_io.is_a?(String)
-        File.open(filename_or_io)
+      if filename_or_io.is_a?(String)
+        initialize_with_filename(filename_or_io)
       elsif filename_or_io.is_a?(IO)
-        filename_or_io
+        initialize_with_io(filename_or_io)
       else
         raise TypeError.new(filename_or_io.class)
       end
       check_for_classid_and_version()
     end
 
+    def initialize_with_filename(filename)
+      initialize_with_io(File.open(filename))
+    end
+
+    def initialize_with_io(io)
+      @io = Zlib::GzipReader.new(io)
+    rescue Zlib::GzipFile::Error
+      @io = io
+      @io.rewind
+    end
+
     def each(&blk)
       while not @io.eof?
         url_size = @io.read(4).unpack("i").first
@@ -19,7 +30,8 @@ module EightyLegs
         data = @io.read(data_size)
         blk.call(Entry.new(url, data))
       end
-      @io.seek(8, IO::SEEK_SET)
+      @io.rewind
+      @io.read(8)
     end
 
     private

data/test/test_eighty_format.rb

@@ -2,10 +2,16 @@ require 'test/helper'
 
 class TestEightyFormat < Test::Unit::TestCase
   TEST_80 = 'test/test_data/wikipedia.80'
+  TEST_80_GZ = 'test/test_data/wikipedia.80.gz'
   TEST_80_HTML = 'test/test_data/wikipedia-index.html'
   
   def test_parse_correct
-    [File.open(TEST_80), TEST_80].each do |file_or_io|
+    [
+     File.open(TEST_80),
+     TEST_80,
+     File.open(TEST_80_GZ),
+     TEST_80_GZ
+    ].each do |file_or_io|
       howmany = 0
       file = EightyFormat.new(file_or_io).each do |entry|
         assert_equal("http://en.wikipedia.org/wiki/Main_Page", entry.url)
@@ -16,7 +22,12 @@ class TestEightyFormat < Test::Unit::TestCase
       assert_equal(1, howmany)
     end
      
-    [File.open(TEST_80), TEST_80].each do |file_or_io|
+    [
+     File.open(TEST_80),
+     TEST_80,
+     File.open(TEST_80_GZ),
+     TEST_80_GZ
+    ].each do |file_or_io|
       howmany = 0
       EightyFormat.new(file_or_io).each do |entry|
         howmany += 1

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification 
 name: 80legs
 version: !ruby/object:Gem::Version 
-  hash: 27
+  hash: 23
   prerelease: 
   segments: 
   - 0
-  - 1
+  - 2
   - 0
-  version: 0.1.0
+  version: 0.2.0
 platform: ruby
 authors: 
 - Jan Szumiec
@@ -15,7 +15,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2011-07-08 00:00:00 +02:00
+date: 2011-07-11 00:00:00 +02:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -117,6 +117,7 @@ files:
 - test/helper.rb
 - test/test_data/wikipedia-index.html
 - test/test_data/wikipedia.80
+- test/test_data/wikipedia.80.gz
 - test/test_eighty_format.rb
 has_rdoc: true
 homepage: http://github.com/jasiek/80legs